We study a simple extension of the decoupling limit of boundary effctive actions for the Dvali-Gabadadze-Porrati model, by covariantizing the π lagrangian and coupling to gravity in the usual way. This extension agrees with DGP to leading order in Mpl^−1 , and simplifies the cosmological analysis. It is also shown to softly break the shift symmetry, while still being consistent with solar system observations. The generally covariant equations of motion for π and the metric are derived, then the cosmology is developed under the Cosmological Principle. Three analytic solutions are found and their stability is studied. Interesting DGP phenomenology is reproduced, and we consider one of the stable solutions. The cosmological analogue of the Vainshtein effect is reproduced and the effective equation of state, w_π, is shown to be bounded by −1 from above. This solution is additionally shown to be an attractor solution in an expanding universe. We evolve π numerically and reproduce these properties, and show that the universe will go through a contraction phase, due to this π field. We then place a constraint on r_c ≥ 10^29 cm, given recent WMAP5 data. This lower bound on r_c gives an upper bound on the anomalous perihelion precession of the moon ∼ 1 × 10^−13, 2 orders of magnitude below current experimental precision.

We study a simple extension of the decoupling limit of boundary effctive actions for the Dvali-Gabadadze-Porrati model, by covariantizing the π lagrangian and coupling to gravity in the usual way. This extension agrees with DGP to leading order in Mpl^−1 , and simplifies the cosmological analysis. It is also shown to softly break the shift symmetry, while still being consistent with solar system observations. The generally covariant equations of motion for π and the metric are derived, then the cosmology is developed under the Cosmological Principle. Three analytic solutions are found and their stability is studied. Interesting DGP phenomenology is reproduced, and we consider one of the stable solutions. The cosmological analogue of the Vainshtein effect is reproduced and the effective equation of state, w_π, is shown to be bounded by −1 from above. This solution is additionally shown to be an attractor solution in an expanding universe. We evolve π numerically and reproduce these properties, and show that the universe will go through a contraction phase, due to this π field. We then place a constraint on r_c ≥ 10^29 cm, given recent WMAP5 data. This lower bound on r_c gives an upper bound on the anomalous perihelion precession of the moon ∼ 1 × 10^−13, 2 orders of magnitude below current experimental precision.

While general relativity is a very successful theory of gravity, having thus far passed all observational tests with flying colors, it is thought to be incomplete. Indeed, we lack an ultimate high energy theory in which general relativity and quantum mechanics are both valid. We consider extensions to the action of general relativity, and seek to place constraints on these alternative theories using astrophysical tests. General relativity has been extensively tested in the solar system, but not with precision in strong gravity systems. We discuss constraints that could be placed on alternative theories using neutron stars. We find that we may not be able to distinguish between general relativity and some alternative theories in the spacetimes around black holes. We also discuss constraints from cosmological tests, and show that instabilities can appear.Adding higher-order terms to the action of general relativity can introduce new dynamical degrees of freedom and instabilities. From the standpoint of effective field theory however, these alternative theories are inconsistent because they are not unitary. In an effective field theory, no new degree of freedom is introduced. This also means that extra polarizations of gravitational waves, which are predicted by some alternative theories, would not be present in an effective field theory.We then consider an effective field theoretic formulation for gravitational radiation called Non-Relativistic General Relativity (NRGR). We study the gravitational wave emission in non-relativistic coalescing compact binaries, which are thought to be powerful emitters of gravitational waves. While NRGR is based on the post-newtonian (PN) approximation to general relativity, and should therefore be in complete agreement with other post-newtonian methods, the effective field theory approach provides two major advantages: it provides a consistent framework for the dynamics using a lagrangian formulation; also, one can in principle compute observables to all orders in the orbital velocity in a systematic way. We provide a brief overview of NRGR methods, and present the NRGR calculation of the subleading spin-orbit correction to the newtonian potential.

Durant cette thèse, nous avons exploré l'impact qu'engendre une modification de la gravitation sur les courants de marée. Nous avons effectuer les premières simulations N-corps de la formation d'un courant de marée galactique dans le paradigme MOND. Nous avons comparer les résultats ainsi obtenus aux prédictions du modèle standard de la cosmologie, ΛCMD, dans le but de trouver des différences entres-elles qui pourraient être observables. Nous avons remarqué que la brisure du principe d'équivalence fort engendrée une telle modification de la gravitation conduisait les amas globulaires à avoir une morphologie ovoïdale, contrairement à la dynamique Newtonienne où ceux-ci sont elliptiques. Cette morphologie des amas génère également une asymétrie de longueur et de nombre entre les deux bras d'un courant de marée similaire à celle observée récemment dans le courant de Palomar 5. / During this thesis, we explored the impact of a modification of the gravitation on the tidal streams. We made the first N-body simulations of the formation of a galactic tidal stream in the MOND paradigm. We compared the results obtained with the predictions of the standard model of the cosmology, ΛCMD, with the aim of finding differences between them that can be observed. We have noticed that the break of the strong equivalence principle generated by a such modification of the gravitation led to a lopsided morphology of the globular clusters, contrary to their elliptical shape in Newtonian dynamics. This morphology of the cluster also generates an asymmetry of length and of number between the two arms of a tidal stream similar to that observed recently in the Palomar 5 stream.

When recent observational data and the GR+FRW+CDM model are combined we obtain the result that the Universe is accelerating, where the acceleration is due to some not-yet-understood "dark sector". There has been a considerable number of theoretical models constructed in an attempt to provide a description of the dark sector: dark energy and modified gravity theories. The proliferation of modified gravity and dark energy models has brought to light the need to construct a "generic" way to parameterize perturbations in the dark sector. In this thesis we discuss our new way of approaching this problem. We write down an effective action for linearized perturbations to the gravitational field equations for a given field content and use it to compute generalized gravitational field equations for linearized perturbations. Our approach is inspired by that taken in particle physics, where the most general modifications to the standard model are written down for a given field content that is compatible with some assumed symmetry (which we take to be isotropy of the background spatial sections). After applying the formalism we obtain equations of state for dark sector perturbations, where the number of free parameters for wide classes of theories are identified.

No description available.

La cosmologie a atteint une ère passionnante. Pour la première fois dans l’histoire humaine, un modèle quantitatif pour la formation et l’évolution de l’Univers existe,expliquant une gamme très variée de phénomènes et ayant été testé avec une impressionnante. De plus, nous sommes à l’aube d’une époque dans laquelle nous aurons à notre disposition une abondance de données de grande qualité, ce qui nous permettra d’utiliser la cosmologie comme un outil pour tester la physique fondamentale.En particulier, comme les structures de grandes échelles de l’Univers sont gouvernées par la force de gravité, la cosmologie peut être utilisée pour tester la théorie de la relativité générale d’Einstein. Cette idée a inspiré la plupart des travaux décrits dans ce manuscrit, dans lequel j’ai étudié des théories alternatives au modèle standard de la cosmologie et des tests à grandes échelles pour la relativité générale.Dans la première partie de ma thèse, je me suis concentré sur les “théories tenseur-scalaire” de la gravité. Ce sont des théories alternatives de la gravité, dans lesquelles un champ scalaire, qui est responsable de l’accélération de l’expansion de l’Univers, est ajouté à l’action de Einstein-Hilbert. Dans le deuxième chapitre, j’ai décrit le modèle de K-mouflage, une “théorie tenseur scalaire”dans laquelle le champ scalaire possède un terme cinétique non-standard, en montrant son effet non négligeable sur la dynamique des amas des galaxies. J’ai aussi montré comment cet effet peut être utilisé pour contraindre le modèle de “K-mouflage” en utilisant des observations en rayon X.En particulier, j’ai montré que le cisaillement cosmique a un pouvoir assez limité actuellement pour contraindre ces théories, à cause de la faible précision des observations actuelles et des dégénérescences avec les processus baryoniques.Dans le cinquième chapitre, j’ai donné une description de mon étude des relations de cohérence. Ce sont des relations entreles fonctions de corrélation des champs cosmiques à (n + 1) et n points, valables aussi dans le régime non-linéaire.Leur intérêt vient du fait que leur dérivation dépend seulement du Principe d’Équivalence et de l’hypothèse de conditions initiales Gaussiennes, et donc elles peuvent être utilisées pour tester ces hypothèses fondamentales du modèle standard de la cosmologie.Des relations similaires ont été étudiées auparavant, mais j’ai montré comment il est possible d’obtenir des relations qui ne s’annulent pas lorsque tous les champs sont considérés au même instant. J’ai utilisé ce résultat pour obtenir des relations de cohérence entre fonctions de corrélation de quantités observables, notamment le champ de densité des galaxies et la fluctuation de température du fond diffus cosmologie donnée par l’effet Sachs-Wolfe. Ces relations peuvent être utilisées pour des tests de la relativité générale,reposant sur des observations par satellites, sans avoir besoin de modéliser la physique des baryons aux petites échelles.Enfin, j’ai donné un aperçu des quelques perspectives possibles pour poursuivre le travail décrit dans ce manuscrit.En particulier, j’ai souligné comment des simulations numérique sont nécessaires pourmieux comprendre la formation des structuresdans le contexte des modèles “K-mouflage” et“ultra-local”. En outre, elles peuvent être aussiutilisées pour tester les hypothèses sous-tendantl’analyse des lentilles gravitationnelles faiblesprésentée dans ce manuscrit, surtout pourdistinguer l’effet de la physique des baryons etdes neutrinos de l’effet des théories de gravitémodifiée sur le cisaillement cosmique. En cequi concerne les relations de cohérence, uneétude de faisabilité de leur mesure avec les“surveys” cosmologiques est nécessaire, pourcomprendre si elles peuvent donner descontraintes compétitives sur les théoriesalternatives de la gravité. / The study of physical cosmology has reached an exciting era. For the first time in human history, a quantitative model for the formation and the evolution of the Universe exists, which explains a wide range of phenomena and has been tested with incredible accuracy during the last century. Moreover, weare approaching a time when a bounty of high quality cosmological data will be available,allowing us to use cosmology as a tool to test fundamental physics.In particular, as the large-scale structures of the Universe are governed by gravity, cosmology can help us to asses the correctness of Albert Einstein’s general relativity. This idea fueledmost of the work described in this manuscript,in which we study alternative theories to the standard cosmological model and large-scale structure tests for general relativity.In particular, we focus on two scalar-tensor theories of gravity, the K-mouflage models described in Chapter 2 and the ultra-localmodels of gravity presented in Chapter 3. The K-mouflage theories are k-essence models with a non-standard kinetic term that were already studied at the linear and background levels. In this manuscript, we extend this description showing how the scalar field, which is responsible for the late time acceleration of the Universe, has a non-negligible impact on cluster dynamics, arguing that future surveysmay have the power of constraining K-mouflage models via X-ray observations. Next,we study the ultra-local models of gravity,where a scalar field with a negligible kinetic term is added to the Einstein-Hilbert action,investigating how the latter modifiescosmological structure formation and howthese models can be related to super-chameleonmodels.In Chapter 4, we present a cosmic shear data analysis in the context of f (R) and Dilaton models, and we show how current data canaccommodate most of the theories considered,once baryonic and neutrino physicsdegeneracies are taken into account. Finally, in Chapter 5 we present an analysis of consistency relations for large-scale structures,which are non-perturbative relations among correlations of cosmic fields. They are alsovalid in the non-linear regime, where very few analytical results are known, and only rely on the Equivalence Principle and primordial Gaussian initial conditions. This makes them a powerful tool to test general relativity and inflationary models using the cosmologic allarge-scale structures.We provide here the first non-vanishing equaltime consistency relations, which we use to obtain consistency relations involving observational quantities.In this Thesis manuscript, we highlight the major results of the full analysis done in the articles that are appended to the main text,where the reader can quench any thirst for technical details.

Vingt années se sont écoulées depuis la découverte de l'expansion accélérée de l'Univers, ravivant l'intérêt pour les théories alternatives de la gravité. Ajouter un champ scalaire à la métrique habituelle de la relativité générale est l'une des manières les plus simples de modifier notre théorie de la gravité. En parallèle, nos connaissances sur les trous noirs et les étoiles à neutrons sont en plein essor, grâce notamment au développement de l'astronomie par ondes gravitationnelles. Cette thèse se situe au carrefour entre les deux domaines : elle étudie les propriétés des objets compacts dans les théories tenseur-scalaire généralisées. Je commence par rappeler les théorèmes d'unicité essentiels établis depuis les années soixante-dix. Après avoir présenté le théorème d'unicité pour les trous noirs en théorie de Horndeski, je l'étends aux étoiles. La deuxième partie de cette thèse détaille les différentes manières de contourner ce théorème. Parmi elles, je présente des solutions où la dépendance temporelle du champ scalaire permet de le raccorder à une solution cosmologique, mais aussi des trous noirs statiques et asymptotiquement plats. Dans la troisième partie, j'établis un critère important pour la stabilité de ces solutions, qui s'appuie sur leur structure causale. C'est aussi l'occasion d'étudier la propagation des ondes gravitationnelles au voisinage de trous noirs, et de sélectionner les théories dans lesquelles les ondes gravitationnelles se propagent à la même vitesse que la lumière. / Twenty years have passed since the discovery of the accelerated expansion of the Universe, reviving the interest for alternative theories of gravity. Adding a scalar degree of freedom to the usual metric of general relativity is one of the simplest ways to modify our gravitational theory. In parallel, our knowledge about black holes and neutron stars is booming, notably thanks to the advent of gravitational wave astronomy. This thesis is at the crossroads between the two fields, investigating the properties of compact objects in extended scalar-tensor theories. I start by reviewing essential no-hair results established since the seventies. After discussing the no-hair theorem proposed for black holes in Horndeski theory, I present its extension to stars. The second part of the thesis investigates in detail the various ways to circumvent this theorem. These notably include solutions with a time-dependent scalar field in order to match cosmological evolution, but also static and asymptotically flat configurations. In a third part, I establish an important stability criterion for these solutions, based on their causal structure. It is also the occasion to study the propagation of gravitational waves in black hole environments, and to select the theories where gravitational waves travel at the same speed as light.

京都大学 / 0048 / 新制・課程博士 / 博士(理学) / 甲第20656号 / 理博第4321号 / 新制||理||1621(附属図書館) / 京都大学大学院理学研究科物理学・宇宙物理学専攻 / (主査)教授 佐々木 節, 教授 田中 貴浩, 教授 川合 光 / 学位規則第4条第1項該当 / Doctor of Science / Kyoto University / DFAM

We introduce an elastic constitutive model of gravity that enables the interpretation of cosmological observations in terms of established ideas from Solid Mechanics and multiscale modeling. The behavior of physical space is identified with that of a material-like medium called "cosmic fabric," which exhibits constitutive behavior. This cosmic fabric is a solid hyperplate that is broad in the three ordinary spatial dimensions and thin in a fourth hyperspatial dimension. Matter in space is treated as fabric inclusions that prescribe in-plane (three-dimensional) strain causing the transverse bending of the fabric into the fourth hyperspatial dimension. The linearized Einstein-Hilbert action, which governs the dynamics of physical space, is derived from postulating Hooke’s Law for the fabric, and the Schwarzschild metric is recovered from investigating matterabric interactions. At the continuum length scale, the Principle of Relativity is shown to apply for both moving and stationary observers alike, so that the fabric’s rest reference frame remains observationally indistinguishable at such a length scale. Within the Cosmic Fabric paradigm, the structural properties of space at different hierarchical length scales can be investigated using theoretical notions and computational tools from solid mechanics to address outstanding problems in cosmology and fundamental physics. For example, we propose and offer theoretical support for the "Inherent Structure Hypothesis", which states that the gravitational anomalies currently attributed to dark matter may in fact be manifestations of the inherent (undeformed) curvature of space. In addition, we develop a numerical framework wherein one can perform numerical "experiments" to investigate the implications of said hypothesis.